Jiajun Wu

TL;DR: This work proposes a new approach to learn an interpretable distributed representation of scenes, using a deterministic rendering function as the decoder and a object proposal based encoder that is trained by minimizing both the supervised prediction and the unsupervised reconstruction errors.

TL;DR: In this article, a new generative model, Visual Object Networks (VON), is proposed to synthesize natural images of objects with a disentangled 3D representation by unraveling the image formation process into three conditionally independent factors (shape, viewpoint, and texture).

TL;DR: In this paper, an approach to audio-visual embodied navigation that takes advantage of both visual and audio pieces of evidence is presented. But it is based on three key ideas: a visual perception mapper module that constructs its spatial memory of the environment, a sound perception module that infers the relative location of the sound source from the agent, and a dynamic path planner that plans a sequence of actions based on the audio visual observations and the spatial memory in the environment to navigate toward the goal.

TL;DR: The proposed ShapeHD pushes the limit of single-view shape completion and reconstruction by integrating deep generative models with adversarially learned shape priors, penalizing the model only if its output is unrealistic, not if it deviates from the ground truth.

TL;DR: In this article, the authors propose ShapeHD, which integrates deep generative models with adversarially learned shape priors, which serve as a regularizer, penalizing the model only if its output is unrealistic, not if it deviates from the ground truth.